3 ms·
Why not evaluate (estimate) the expected value of sample information of the different experiments, and compare the costs of the different experiments to the EVS
by drdeca 6y ago
Why not evaluate (estimate) the expected value of sample information of the different experiments, and compare the costs of the different experiments to the EVSIs?
Is it too difficult to estimate the EVSI ?
- rowanG077 6y agoAny metric you can imagine is nothing more then a guess. That's why it's research. The researchers guess and tests. Some guesses can be observably better but that's not quantifiable. At least that's what I think.
- newsclues 6y agoSurely you see the hazard in funding expensive wildly speculative projects? If you fail once, fail again but bigger... who cares it’s not our money and the bigger our project the more we pay ourselves...
- rowanG077 6y agoThat's why scientific projects are still judged by panels. It's not like money is thrown at any random project. It must have merit and be judged to actually be able to test a hypothesis.